symbolic inference

symbolic inference

The derivation of new facts from known facts and inference rules. This is one of the fundamental operations of artificial intelligence and logic programming languages like Prolog.

Inference is a basic part of human reasoning. For example given that all men are mortal and that Socrates is a man, it is a trivial step to infer that Socrates is mortal. We might express these symbolically:

man(X) => mortal(X). man(socrates).

("if X is a man then X is mortal" and "Socrates is a man"). Here, "man", "mortal" and "socrates" are just arbitrary symbols which the computer manipulates without reference to or knowledge of their external meaning. A forward chaining system (a production system) could use these to infer the new fact

mortal(socrates).

simply by matching the left-hand-side of the implication against the fact and substituting socrates for the variable X.
References in periodicals archive ?
Of special interest was the idea of solving problems traditionally considered within the realm of numerical approximation by means of symbolic inference. This idea was illustrated by Steven Diamond (Stanford) in a keynote discussing convex optimization with abstract (symbolic) linear operators.
Langer outlines modes of abstraction and forms of symbolic inference that operate prior to (or beneath) any verbal description.
"Symbolic Inference; or, Kenneth Burke and Ideological Analysis." Ideologies of Theory, Volume 1.

Full browser ?